United KingdomFinance and insurance2025 – ongoingRBC Brewin Dolphin
Generative AI preparation of annual client review packs for financial planners
RBC Brewin Dolphin saved about 20 minutes per client meeting pack on average after putting generative AI on review preparation
Problem
Financial planners at RBC Brewin Dolphin prepared about 14,000 annual client reviews a year. They gathered information by hand from five separate source systems before each meeting.
Approach
The firm built a generative AI workflow on the Databricks platform. Advisors upload documents, and the system parses them and extracts insights. It then builds a meeting report using retrieval-augmented generation, with 30 to 40 tailored prompts per session, mainly on Llama 3.1 and 3.3 models. Natural-language interfaces and fine-tuned embedding models were planned next.
Results
Takeaway
The firm gave the 20-minute saving per pack as an average. The yearly hours, the automation share and the cost cut are estimates for a rollout still under way.
Categories
- Size
- Large enterprise
- Industry
- Finance and insurance
- Business function
- Internal work and knowledge management
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how RBC Brewin Dolphin built a generative AI workflow to prepare annual client review packs across its 33 UK and Ireland offices. It reports about 20 minutes saved per pack and 4,700 hours of potential annual savings. It also gives projections of 80-90% automation of pack preparation and an estimated 50% cut in administrative costs.
Full details are in the original sources.